基于数理统计方法的闪电气候研究

基于广东省1999—2015年闪电定位系统数据,利用数理统计方法得出广东省闪电气候年际变化特征、季节变化特征、月变化特征,利用最小二乘法得到闪电密度气候变化趋势系数和年、季节的气候倾向率。利用Matlab来对广东省1999—2013年的雷暴日进行分析,得出其年际周期变化规律,最后采用EOF方法,借助Arc GIS平台得出广东省闪电密度空间向量场分布图。结果表明:由1999—2015年,广东全省的地闪密度总体趋势是逐渐增加,闪电日数最多的月份集中在6—8月(夏季);雷暴日存在变化具有周期性、规律性,分别有13～16 a,7～11 a、2～6 a 3种不同尺度的周期;闪电密度空间分布特征主要有一致型、局部型、纬向型和经向型4种类型。     

Factors controlling the seasonality of the Madden-Julian Oscillation

By analyzing observational data, previous studies have indicated that the tropical Madden-Julian Oscillation (MJO) is active during the boreal winter but relatively weak during the boreal summer. However, the factors that control seasonal MJO variation are not clear. To quantitatively understand the relative contributions of the occurrence frequency of enhanced MJO events and their averaged strength and lifespan to seasonal MJO amplitude, we defined the MJO events of 1979–2014 and analyzed their features in different seasons by using the Real-time Multivariate MJO (RMM) index and the newly proposed RMM-r index. The results indicate that the MJO events show a higher frequency of occurrence, a stronger intensity and a longer duration during the boreal winter (Dec.–Feb.) and spring (Mar.–May). However, the frequency, strength and lifespan of MJO events are all reduced during the boreal summer (Jun.–Aug.) and autumn (Sep.–Nov.). The enhanced MJO events in winter–spring also show a large ratio of variance for eastward to westward components. To elucidate how large-scale background fields affect seasonal MJO variation, a series of sensitivity experiments was conducted by using a 2.5-layer model that can simulate MJO-like features. It is found that the variation in low-level moisture (vertical wind shear) is the key large-scale factor affecting the seasonal variation in MJO strength (in propagation). In comparison with the summer–autumn seasons when the MJO is relatively weakened, the relatively abundant low-level moisture near the equator during boreal winter–spring may strengthen the development of MJO convection and circulation, whereas the relatively weak easterly shear (or the westerly shear anomaly) is conducive to the enhancement of an eastward-propagating MJO component.     

STATISTICAL ANALYSIS OF LOW-LEVEL JET STREAMS IN NANJING AREA BASED ON WIND PROFILER DATA

In order to understand the activity characteristics of low-level jets in the Nanjing area, statistical analysis and comparative study are carried out on their monthly and diurnal variations, characteristics of their cores and accompanying weather conditions using wind profile data in 2005—2008 collected by two wind profilers. The results show that low-level jets have significant monthly and diurnal variations. They occur more frequently in spring and summer than in autumn and winter and are more active in early morning and at night, with the maximum wind speed usually occurring at midnight. The central part of the low-level jet occurs mainly at the height of less than 1400 meters, and the enhancement of central speed is beneficial to the appearance of precipitation. Meanwhile, when the low-level jet appears in summer, it helps cause heavy rain. The statistical results of the boundary wind profiler are well consistent with those of the tropospheric wind profiler. Two kinds of wind profilers also have the capability of continuously detecting the development of low-level jets.     

How many Stakes are Required to Measure the Mass Balance of a Glacier?

Glacier mass balance is estimated for South Cascade Glacier and Maclure Glacier using a one-dimensional regression of mass balance with altitude as an alternative to the traditional approach of contouring mass balance values. One attractive feature of regression is that it can be applied to sparse data sets where contouring is not possible and can provide an objective error of the resulting estimate. Regression methods yielded mass balance values equivalent to contouring methods. The effect of the number of mass balance measurements on the final value for the glacier showed that sample sizes as small as five stakes provided reasonable estimates, although the error estimates were greater than for larger sample sizes. Different spatial patterns of measurement locations showed no appreciable influence on the final value as long as different surface altitudes were intermittently sampled over the altitude range of the glacier. Two different regression equations were examined, a quadratic, and a piecewise linear spline, and comparison of results showed little sensitivity to the type of equation. These results point to the dominant effect of the gradient of mass balance with altitude of alpine glaciers compared to transverse variations. The number of mass balance measurements required to determine the glacier balance appears to be scale invariant for small glaciers and five to ten stakes are sufficient.     

Kalman滤波方法在黑河出山径流年平均流量预报中的应用

应用Kalman滤波方法,以流域平均降水量、气温、太阳黑子相对数及莺落峡站6年显著周期序列等为控制参数,对黑河出山径流年平均流量的预测问题进行了初步研究和探讨。     

An application of the Kerr black hole fly-wheel model to statistical properties of QSOs/AGNs

The aim of this work is to demonstrate the properties of the magnetospheric model around Kerr black holes (BHs), the so-called fly-wheel (rotation driven) model. The fly-wheel engine of the BH–accretion disc system is applied to the statistics of QSOs/AGNs. In the model, the central BH is assumed to be formed at z ∼10² and obtains nearly maximum but finite rotation energy (∼extreme Kerr BH) at the formation stage. The inherently obtained rotation energy of the Kerr BH is released through a magnetohydrodynamic process. This model naturally leads to a finite lifetime of AGN activity.
Nitta, Takahashi & Tomimatsu clarified the individual evolution of the Kerr BH fly-wheel engine, which is parametrized by BH mass, initial Kerr parameter, magnetic field near the horizon and a dimensionless small parameter. We impose a statistical model for the initial mass function (IMF) of an ensemble of BHs using the Press–Schechter formalism. With the help of additional assumptions, we can discuss the evolution of the luminosity function and the spatial number density of QSOs/AGNs.
By comparing with observations , it is found that a somewhat flat IMF and weak dependence of the magnetic field on the BH mass are preferred. The result explains well the decrease of very bright QSOs and decrease of population after z ∼2.     

岩石变形图像分析的计算机统计方法初探

本文利用岩石变形形成的显微构造现象(如压力影、亚颗粒、位错、动态重结晶颗粒等),通过图象分析计算机数理统计方法,来推测地质体变形特征、变形环境、应力状态等。      

基于多元数据分析的我国PM2.5浓度及其主控因子的时空分布特征研究

基于2013~2016年空气质量监测台站资料,利用经验正交分解、功率谱分析、BP典型相关分析等多元数据分析方法解析了中国地区细颗粒物（PM2.5）主要模态的时空特征,并与排放源和气象场建立了相关关系,得到以下结论:中国地区PM2.5场存在两个主要模态,其中第一主模态为一致增加模态,强度中心位于西北地区东部—华北南部地区;其时间序列呈显著下降趋势。第二主模态主要表现为南北反向变化的偶极子型分布,其大值区分别位于华北中南部和长江中下游地区。其中,PM2.5第一模态可以看作平均态,主要受平均排放场和环流场及大地形的影响,在北方的表现更为显著。PM2.5第二模态可看作偏离平均场的一种变化态,在冬季更可能和冷空气活动有关。冷空气的强弱决定了污染累积的位置以及输送的方式,其作用是使得南方的污染明显偏离平均态,故第二主模态在南方的表现更为显著。本研究有效地利用了多元数据分析方法研究了我国大气污染的演变机理,可为进一步认清大气污染的形成规律提供科技支撑。     

卫星遥感信息处理与多元统计分析

本文针对遥感信息的数据特征探讨了卫星遥感信息处理与多元统计算法之间的关系,并从实际出发对最小二乘残差技术,目标模式识别,增强主成份分析等方法的应用作了介绍。     

Predicting New Hampshire indoor radon concentrations from geologic information and other covariates

 Generalized geologic province information and data on house construction were used to predict indoor radon concentrations in New Hampshire (NH). A mixed-effects regression model was used to predict the geometric mean (GM) short-term radon concentrations in 259 NH towns. Bayesian methods were used to avoid over-fitting and to minimize the effects of small sample variation within towns. Data from a random survey of short-term radon measurements, individual residence building characteristics, along with geologic unit information, and average surface radium concentration by town, were variables used in the model. Predicted town GM short-term indoor radon concentrations for detached houses with usable basements range from 34 Bq/m³ (1 pCi/l) to 558 Bq/m³ (15 pCi/l), with uncertainties of about 30%. A geologic province consisting of glacial deposits and marine sediments was associated with significantly elevated radon levels, after adjustment for radium concentration and building type. Validation and interpretation of results are discussed. Received: 20 October 1997 · Accepted: 18 May 1998